The Galaxy S5 does not have temperature or humidity sensors | Hacker News Reader

The Galaxy S5 does not have temperature or humidity sensors

opensignal.com

48 pointscleis49 comments

It's most likely due to the water-proofing. The S4 rugged version has the same limitation.

I talked to an engineer from Sensirion (the makers of the temperature sensor on the S4) a couple of weeks ago. He told me that the waterproofing was the reason that they couldn't include their sensor in the S5.

The S5 barometer apparently survives underwater... I see no reason why a temperature/humidity sensor can't be engineered to.

Conflicting information, I just talked with someone from Sensirion who was quite central to the the SHTC1 project, he confirmed it was NOT the waterproofing that was the issue and pointed out there are Japanese devices with the same chip that are waterproof, even to IP68 (2m depth).

Well that's disappointing. Adding gimmicky(imo) sensors and removing functional/useful sensors.

Obviously Samsung is thinking from a consumer prospective, but still very disappointing.

"Obviously Samsung is thinking from a consumer prospective, but still very disappointing."

Wait, what? Samsung is trying to make majority of customers happy and that's disappointing? This is what you are saying?

Did you miss the "gimmicky" part? Samsung is thinking from the customer perspective in that they are usually brought in by flashy "new" features. But that doesn't mean that these features actually pan out into anything useful.

I guess "consumer perspective" means "optimizing short-term amount of money extracted from customers for value given" instead of "providing the most awesome product we can and reaping benefits a little bit later".

When you give developers more tools, they surprise you with what they come up with. Consumers didn't say they wanted accelerometers, but once developers got ahold of them, they proved massively useful.

If you give me a new sensor, I might not have any ideas on how to make it useful, but if one developer has one good idea, that benefits me.

Exactly correct! Google and Motorola put a barometer into the Xoom to test GPS altitude improvements, and it turned out that some developers (like me, and like the opensignal team) discovered that the sensor actually has world-changing potential in atmosphere modelling. Huge benefit from a simple sensor addition that didn't even really have a clear purpose at launch.

But then that goes back to funkyy's point: there may be lots of unexpected things developers can do with it, but none of them so far are things customers care about. Atmospheric modeling? Really cool! Do customers care? Heck no.

They'll care when they get very high accuracy weather forecasts! New models will result in more accurate, more personalized and relevant forecasts that I think customers will love. This is a very tough problem that nobody has cracked yet, but it looks promising in the near future. We'll see.

But there is still the question of value proposition to the consumer.

Is the incrimental marginal cost of including the sensor low enough that the improvements in weather forcasting will be demananded by consumers?

I dont believe the target demographic of samsung is swayed towards their products by an additional 5 even 10 degrees of weather accuracy. But i have no data to back this up.

Are there any papers on the topic of including this new data written by actual meteorologists?

Yes, there are quite a few relevant papers. Here's a preliminary study from 2006 regarding weather forecasting using surface pressure measurements on the synoptic scale: http://journals.ametsoc.org/doi/abs/10.1175/BAMS-87-2-175

A number of papers have been published since as that work has continued. So we know that surface pressure data is useful on a large scale; is it useful on a micro- or mesoscale? A few researchers are actively writing and publishing papers on this topic. Cliff Mass at the University of Washington, along with his colleagues and students, are leading the charge.

https://ams.confex.com/ams/94Annual/webprogram/Paper236282.h...

Customers are idiots. To wit:

"If I had asked people what they wanted, they would have said faster horses." -- Henry Ford

Customers aren't always idiots. However, there's a special kind of engineering idiocy that thinks that having a humidity sensor in a phone is a great idea.

Even with a large sample, temperature and humidity data is useless unless it is ensured to be properly measured (ie outdoors, away from the person, etc). Pressure data is another story, and although it is still low quality it can be smoothed.

That's why I prefer pressureNET to opensignal. Also pressureNET is actually open source, unlike opensignal which sells all your smartphone/personal data to anyone who wants it.

WeatherSignal is a slightly different project to OpenSignal (but it's by the same people) - we're sharing data with several academic institutions and independent researchers, and will make the feed fully open. Also NB, PressureNet is a great project, WeatherSignal also collects pressure data but I think other sensors are relevant.

We have an algorithm in WeatherSignal that tries to determine whether users are indoors or outdoors - there's a roof icon that appears or disappears, try it, it's pretty accurate during the day time.

WeatherSignal project is basically funded by OpenSignal sales, we sell to carriers and regulators -- who can act on the data to improve service.

(I'm James Robinson, a co-founder of OpenSignal/WeatherSignal)

> and removing functional/useful sensors.

Are temperature and humidity really that useful? I know many Android phones have had them, but have they really be used? Do they really add much to the device?

to the individual they did not pose much benefit, but to society as a whole they were extremely useful. Through programs like pressureNET, literally opening up new avenues for data recording that never existed before.

In the short term these new sensors will provide more benefits to the consumer, but to the majority they will be something you use 1 or 2 times then forget about.

Temperature would be extremely useful... if not for the fact that I typically store my cellphone millimeters away from my body.

If temperature were useful, then humidity would be as well because then you could calculate local humidex.

It would have more to do with the temperature of the CPU and/or battery than anything dealing with the outside temperature.

Why? My phone tends to be warm because of body heat most of the time. Yeah, the shell can sometimes get warm under load, but that’s the exception. I would guess that if not for body heat and with the right placement a temperature sensor could get useful data.

However, with the phone nearly always being transported so close to the body I really don’t see it being useful.

I believe temperature sensors are required to get barometric data to altitude[1]. That's why all altimeter watches also have temperature though its usually useless while wearing since its reading the temp of your wrist instead of the air.

On a second note I produce an app (http://www.ullrlabs.com) for avalanche danger evaluation and knowing your ambient air temperature is one factor which is helpful in that process. Though I haven't yet tried to read that value on android phones for the app.

[1]See pressure altimeter calibration equation: http://en.wikipedia.org/wiki/Altimeter

Your first note is irrelevant, since all modern phones have a (hot) CPU running while the phone is active. It's why phones don't rely on their temperature sensors for ambient readings in the first place even if they have them.

Temperature and humidity seem gimmicky to me (I don't care about atmospheric conditions in my pocket) while fingerprint and heart rate are at least potentially useful.

I've never found a phone temp sensor to be useful, primarily because it tends to be several degrees higher than ambient due to internal components.

I believe that monitoring internal components is exactly why temperature sensors are useful: the device can shut itself off it it becomes too hot. This can happen if the it has been left on the dashboard of a car during the summer, for example. This can help to lengthen the lifetime of the device.

All cellphones (even very old ones) have some sort of temperature sensor. This is required for dealing with components that are sensitive to heat (oscillators for example).

These are usually not accessible to applications since they (by design) do not report the ambient room temperature.

I don't think those sensors are taken out, I think this is with regard to the outside temperature/humidity sensor.

Internal component temperature measurement isn't something you can get rid of, it's critical to the phone.

This is indeed a common use case for temperature sensors in embedded devices, and I very much doubt there isn't one in the S5. Temperature sensors are included in pretty much all inertial sensor dies nowadays, both for offset/drift calibration and thermal shutdown protection. It's more likely that Samsung simply removed application-level access, although I don't know why they would do so.

+1 - mostly it's good for checking I still have a body temperature.

Thinking from a consumer perspective, the temperature and humidity sensors aren't really useful. It's a phone, not a weather station.

I think though, this gets to the heart of the problem with Samsung. They throw a lot of shit at the wall with no care for design or even how the thing is supposed to be lived with.

(FWIW, I think LG, Nokia and HTC are doing decent work in mobile.)

I love how developers of this app allow themselves to collect numerous data about their users' phones, so much they even track their hardware, model numbers and available APIs. But the worst thing is probably that this has became so common that people don't react negatively anymore towards that kind of privacy violation.

The whole point of that API is crowd source collection of data so everyone can benefit.

Sadly they are selling to the devil: http://opensignal.com/insights/

The data included in OpenSignal insights is aggregated both geographically and temporally at a resolution that makes it impossible to identify any individual user. All we show are averaged stats over a monthly or quarterly period for a given area.

Part of our project is to inform users about how carriers are performing, but also to inform carriers how they're performing so that they can improve their service for their users. We use our data to independently regulate the carrier market, from both the supply and demand sides as we think this is the best way to help effect much-needed improvement. We sell to operators so that we don't have to charge consumers or show them annoying in-app ads!

Why do I care that a mobile developer knows my device model and hardware specs? I'd prefer you to know for bug testing and performance metrics.

Makes sense, after all temp sensors don't really belong to phones. What't the point of knowing it's 35C and the humidity of your pocket? They 'd make sense in cars, once (if ever) they get connected.

Weather signal [0] has developed algorithms to correct the internal sensor measurements for things like this, particularly temperature. So it's not entirely useless.

[0] http://weathersignal.com/

Allow me to be skeptical about the reliability of this algorithm. How can you ever know the temperature of the air outside by measuring the temperature of your pocket? Sure, there's a correlation, to some degree. But what if you wear different pants, or put the phone in your bag? I just don't see how that's possible (to a meaningful degree of precision). I would think data from weather stations would be a lot more reliable, and make the temperature sensor obsolete.

Wouldn't I know when I'd do that? For example, civilization has known for decades now how does a thermometer work. It will show correct body temperature only at a few places. If we can somehow isolate the sensor to not to read its own machine temperature but read the ambient temperature, I think it'll be a big thing.

It presumably relies on having multiple phones recording the temperature, averaging out person-to-person variations in attire (thickness of pants, etc.). If those variations are correlated among people for a given day (everyone is wearing thicker pants), it's probably colder, and that would have to be included in the algorithm.

That sounds like an overkill, in this case it's easier for your phone to get your ambient temperature from some wather server around the world.

Yes, but there's a lot fewer weather stations than there are smartphones. So there's lots of potential, if the sensors and algorithms can be calibrated to account for these systematics.

What other cool sensors do mobile phones have these days?

Many phones also have magnetic compasses, so you can get an accurate bearing without needing GPS signals. I would expect significant advances in inertial navigation systems in mobile devices in the coming years.

https://en.wikipedia.org/wiki/Inertial_navigation_system

The best tl;dr ever!

Reply on news.ycombinator.com